1. Field of the Invention
The present invention relates to the art of electric circuits. Meanwhile, the invention belongs to a technical field of a semiconductor device having an electric circuit as represented by a source-follower circuit, a differential amplifier circuit, a sense amplifier and an operational amplifier, a signal-line drive circuit and a photoelectric converter element.
2. Description of the Related Art
The integrated circuit (IC), for broad use recently on a cellular phone or personal digital assistant, is formed with transistors or resistors as many as several hundreds of thousands to several millions on a silicon substrate in a size of nearly a 5-mm square. This plays an important role in device miniaturization and reliability improvement, and device mass production.
In designing an electric circuit for use on an integrated circuit (IC) or the like, it is frequent cases to design an amplifier circuit having a function to amplify a voltage or current of a signal small in amplitude. The amplifier circuit is broadly used because of a circuit requisite for eliminating strain occurrence to stably operate an electric circuit.
Herein, explained is the configuration and operation of a source-follower circuit, as one example of amplifier circuit. At first, a configuration example of source-follower circuit will be shown in FIG. 5A to explain an operation in a steady state. Next, an operating point of the source-follower circuit will be explained, by using FIGS. 5B and 5C. Finally, an example of source-follower circuit different in configuration from FIG. 5A will be shown in FIGS. 6A and 6B, to explain an operation in a transient state.
At first, a steady state operation is explained by using a source-follower circuit in FIG. 5A.
In FIG. 5A, 11 is an n-channel amplifier transistor while 12 is an n-channel bias transistor. Note that, although the amplifier transistor 11 and bias transistor 12 in FIG. 5A is of an n-channel type, configuration may be by the use of p-channel transistors. Herein, the amplifier transistor 11 and the bias transistor 12 are assumably the same in characteristic and size, for simplification sake. It is further assumed that the current characteristic of them is ideal. Namely, it is supposed that, even if the amplifier transistor 11 or bias transistor 12 is changed in its source-to-drain voltage, there is no change in saturation-region current value.
Meanwhile, the amplifier transistor 11 has a drain region connected to a power line 13 and a source region connected to a drain region of the bias transistor 12. The bias transistor 12 has a source region connected to a power line 14.
The gate electrode of the bias transistor 12 is applied by a bias potential Vb. A power-source potential (high potential power) Vdd is applied onto the power line 13 while a ground potential (low potential power) Vss (=0V) is applied onto the power line 14.
In the source-follower circuit of FIG. 5A, the gate electrode of the amplifier transistor 11 is made as an input terminal so that an input potential Vin can be inputted to the gate electrode of the amplifier transistor 11. Also, the source region of the amplifier transistor 11 is made as an output terminal so that the potential on the source region of the amplifier transistor 11 provides an output potential Vout. The gate electrode of the bias transistor 12 is applied by a bias voltage Vb. When the bias transistor 12 operates in a saturation region, a current denoted by Ib assumably flows. At this time, because the amplifier transistor 11 and the bias transistor 12 are in a series connection, the same amount of current flows through the both transistors. Namely, when a current Ib flows through the bias transistor 12, a current Ib flows also through the amplifier transistor 11.
Herein, determined is an output potential Vout in the source-follower circuit. The output potential Vout is lower in value than the input potential Vin, by an amount of the gate-to-source voltage Vgs1 of the amplifier transistor 11. At this time, the input potential Vin, the output potential Vout and the gate-to-source voltage Vgs1 have a relationship satisfying the following Equation (1).Vout=Vin−Vgs1  (1)
In the case the amplifier transistor 11 is operating in the saturation region, in order to flow a current Ib through the amplifier transistor 11 there is a necessity that the gate-to-source voltage Vgs1 of the amplifier transistor 11 is equal to a bias potential Vb (gate-to-source voltage of the bias transistor 12). If so, the following Equation (2) is held. However, Equation (2) is held only when the amplifier transistor 11 and the bias transistor 12 operate in the saturation region.Vout=Vin−Vb  (2)
Next explained is an operating point of the source-follower circuit by using FIGS. 5B and 5C showing a relationship of between a voltage and a current of the amplifier transistor 11 and bias transistor 12. More specifically, explanation is made on a case that the gate-to-source voltage Vgs1 of the amplifier transistor 11 is same in value as the gate-to-source voltage Vgs2 of the bias transistor 12, by using FIG. 5B. Next explained is a case that the gate-to-source voltage Vgs1 of the amplifier transistor 11 is different in value from the gate-to-source voltage Vgs2 of the bias transistor 12 wherein, for example, the bias transistor 12 is operating in a linear region, by using FIG. 5C.
In FIG. 5B, the dotted line 21 shows a relationship between a voltage and a current when the amplifier transistor 11 has a gate-to-source voltage Vgs1 of Vb. The solid line 22 shows a relationship between a voltage and a current when the bias transistor 12 has a gate-to-source voltage Vgs2 of Vb. Meanwhile, in FIG. 5C, the dotted line 21 shows a relationship between a voltage and a current when the amplifier transistor 11 has a gate-to-source voltage Vgs1 of Vb. The solid line 22 shows a relationship between a voltage and a current when the bias transistor 12 has a gate-to-source voltage Vgs2 of Vb′.
In FIG. 5B, the gate-to-source voltage Vgs1 of the amplifier transistor 11 and the gate-to-source voltage Vgs2 of the bias transistor 12 are in the same value, and further the bias potential Vb and the gate-to-source voltage Vgs2 of bias transistor 12 are in the same value. Consequently, the gate-to-source voltage Vgs1 of the amplifier transistor 11 is in the same value as the bias potential Vb. Namely, this results in Vgs1=Vgs2=Vb. The amplifier transistor 11 and the bias transistor 12 are operating in the saturation region, as shown in FIG. 5B. At this time, the input potential Vin and the output potential Vout have a relationship in a linear form.
On the other hand, in FIG. 5C, the gate-to-source voltage Vgs1 of the amplifier transistor 11 is in a value different from the gate-to-source voltage Vgs2 of bias transistor 12. Furthermore, the gate-to-source voltage Vgs2 of bias transistor 12 is in a same value as the bias voltage Vb. Meanwhile, it is assumed that the gate-to-source voltage Vgs1 of the amplifier transistor 11 is at the bias voltage Vb′. Namely, this results in Vgs2=Vb and Vgs1=Vb′. As shown n FIG. 5C, the amplifier transistor 11 is operating in the saturation region while the bias transistor 12 is operating in the linear region. At this time, the input potential Vin, the output potential Vout and the bias potential Vb′ have a relationship satisfying the following Equation (3).Vout=Vin−Vb′  (3)
Provided that the current flowing upon operating of the bias transistor 12 in the linear region is taken Ib′, Ib′<Ib is given. Namely, by having Vb′<Vb, the both values of the input potential Vin and current Ib′ decrease. Thereupon, the bias potential Vb′ also decreases. At this time, the input potential Vin and the output potential Vout have a non-linear relationship.
Summarizing the above, in order to increase the amplitude of the output potential Vout in the source-follower circuit in a steady state, it is preferred to decrease the bias potential Vb. This is because of the following two reasons.
The first reason is that the output potential Vout can be increased at a small bias potential Vb, as shown in Equation (2). The second reason is that, in the case of a great bias potential Vb value, the bias transistor 12 readily operate in the linear region at a decreased input potential Vin. In case the bias transistor 12 operates in the linear region, the input potential Vin and the output potential Vout are ready to have a non-linear relationship.
Incidentally, because the bias transistor 12 is required in a conduction state, there is a need to provide a greater value of bias potential Vb than a threshold voltage of the bias transistor 12.
So far explained was the operation in a steady state of the source-follower circuit. Subsequently, explanation is made on the operation of the source-follower circuit in a transient state, by using FIGS. 6A and 6B.
The source-follower circuit shown in FIGS. 6A and 6B has a configuration designed by adding a capacitance element 15 to the circuit of FIG. 5A. The capacitance element 15 has one terminal connected to the source region of the amplifier transistor 11 and the other terminal connected to the power line 16. A ground potential Vss is applied onto the power line 16.
The capacitance element 15 has a same potential difference at between its both electrodes as the output potential Vout of the source-follower circuit. Herein, explained is the operation in a case of Vout<Vin−Vb, by using FIG. 6A. Next explained is the operation in a case of Vout>Vin−Vb, by using FIG. 6B.
At first, explanation is made on the operation in a transient state of the source-follower circuit in the case of Vout<Vin−Vb, by using FIG. 6A.
In FIG. 6A, when t=0, the gate-to-source voltage Vgs1 of the amplifier transistor 11 has a greater value than the gate-to-source voltage Vgs2 of the bias transistor 12. Consequently, a great current flows through the amplifier transistor 11 to promptly hold charge on the capacitance element 15. Thereupon, the output potential Vout increases to decrease the gate-to-source voltage Vgs1 value of the amplifier transistor 11.
As time elapses (t=t1, t1>0), the amplifier transistor 11 goes into a steady state when its gate-to-source voltage Vgs1 becomes equal to the bias potential V. At this time, the output potential Vout, the input potential Vin and the bias potential Vb have a relationship satisfying the foregoing Equation (2).
Summarizing the above, in the case of Vout<Vin−Vb, the gate-to-source voltage Vgs1 of the amplifier transistor 11 is greater in value than the bias potential Vb. Accordingly, a great current flows through the amplifier transistor 11, to promptly hold charge on the capacitance element 15. Hence, the time may be short that is required for the capacitance element 15 to hold predetermined charge, in other words the time required in writing a signal to the capacitance element 15.
Next, explanation is made on the operation in a transient state of the source-follower circuit in the case of Vout>Vin−Vb, by using FIG. 6B.
In FIG. 6B, when t=0, the gate-to-source voltage Vgs1 of the amplifier transistor 11 has a smaller value than the threshold voltage of the amplifier transistor 11. Consequently, the amplifier transistor 11 is in a non-conduction state. The charge stored on the capacitance element 15 flows in a direction toward the ground potential Vss through the bias transistor 12, finally being discharged. At this time, because the gate-to-source voltage Vgs2 of the bias transistor 12 is in the same value as the bias potential Vb, the current flowing through the bias transistor 12 is Ib.
As time elapses (t=t1, t1>0), the output potential Vout decreases while the gate-to-source voltage Vgs1 of the amplifier transistor 11 increases. When the gate-to-source voltage Vgs1 of the amplifier transistor 11 becomes equal to the bias potential Vb, a steady state is entered. At this time, the output potential Vout, the input potential Vin and the bias potential Vb have a relationship satisfying the foregoing Equation (2). Note that, in the steady state, the output potential Vout is kept at a constant value, and charge does not flow to the capacitance element 15. Thus, a current Ib flows through the amplifier transistor 11 and bias transistor 12.
Summarizing the above, in the case of Vout>Vin−Vb, the time for the capacitance element 15 to hold predetermined charge, in other words the write time of a signal to the capacitance element 15, relies upon the current Ib flowing through the bias transistor 12. The current Ib relies upon a magnitude of the bias potential Vb. Accordingly, in order to increase the current Ib and shorten the write time of a signal to the capacitance element 15, a necessity is raised to increase the bias potential Vb.
Incidentally, as a method of correcting for threshold-voltage variation of a transistor, there is a method that variation is observed by an output of a circuit a signal has been inputted and thereafter the variation is inputted and fed back thereby carrying out a correction (e.g. see Non-Patent Document 1).    [Non-Patent Document] H. Sekine et al, “Amplifier Compensation Method for a Poly-Si TFT LCLV with an Integrated Data-Driver”, IDRC '97, p. 45-48.
The foregoing operation of the source-follower circuit is to be carried out on an assumption the amplifier transistor 11 and the bias transistor 12 have the same characteristic. However, for the both transistors, variation occurs in the threshold voltage or mobility due to gathering of the factors, such as of variation in gate length (L), gate width (W) and gate insulating film thickness or variation in channel-region crystal state caused due to the difference in fabrication process or substrate used.
For example, it is assumed, in FIG. 5A, that there is variation of 1 V provided that the amplifier transistor 11 has a threshold of 3 V and the bias transistor 12 has a threshold of 4 V. If so, in order to flow a current Ib, there is a need to apply a voltage for the gate-to-source voltage Vgs1 of the amplifier transistor 11 lower by 1 V than the gate-to-source voltage Vgs2 of the bias transistor 12. Namely, Vgs1=Vb−1 results. If so, Vout=Vin−Vgs1=−Vb+1 results. Namely, in case variation occurs even by 1 V in the threshold voltage of the amplifier transistor 11 and bias transistor 12, variation is also caused in the output potential Vout.
The present invention has been made in view of the above problems. It is a problem to provide an electric circuit suppressing against the affection of transistor characteristic variation. More specifically, it is a problem, in an electric circuit having a function of current amplification, to provide an electric circuit capable of supplying a desired voltage while suppressing against the affection of transistor characteristic variation.